Kaixian Chen

This study shows that resistance to different WRN inhibitors in MSI-H cancers creates orthogonal vulnerabilities, enabling alternating therapy and mutation-tolerant inhibitor design.

PBCNet2.0 predicts protein-ligand binding affinity using a cartesian tensor-based Siamese Neural Network, achieving high accuracy and efficiency compared to physics-based simulations.

This study develops AOMP, a graph neural network for predicting AOX metabolism, outperforming benchmarks and identifying new scaffolds validated experimentally.

The paper introduces a Siamese spectral-based graph convolutional network to infer protein targets of chemical compounds from gene expression data, achieving better performance than existing methods.